The invention relates to proteins involved in the prevention of programmed cell death, namely programmed cell death antagonist (PCDA) proteins.
Cells become specified during development through sequential restriction of their potential fates. This process includes mechanisms that monitor differentiation to eliminate, by programmed cell death, cells that have inappropriate specificity or developmental capacity, or that are extraneous (Glucksmann, Biol Rev. 26:59-86 (1951); Saunders, Science 154:604-612 (1966)). Many aspects of tissue development rely on cell death for the selection of proper sets of cells. For example, in vertebrates, massive numbers of neurons generated in early development become eliminated in late stage refinement of connections (Hamburger et al., J. Exp. Zool. 111:457-502 (1949); Oppenheim, Ann. Rev. Neurosci. 14:453-501 (1991)). The mechanism is thought primarily to occur through competition for trophic agents derived from target tissues, which may reinforce appropriate patterns of innervation. Similarly, in the immune system, progenitor cells must generate a great diversity of cell types. The differentiation process relies heavily on regulated cell death to eliminate large numbers of cells of inappropriate reactivity (Fesus, Immunol. Lett. 30:277-282 (1991); Goldstein et al., Immunol. Rev. 121:29-65 (1991)).
The mechanisms and regulation of programmed cell death have a number of implications. For example, the regulation of programmed cell death has implications for oncogenesis (Williams, Cell 65 1097-1098 (1991)), immune disease (Ameisen et al., Immunol. Today 12:102-105 (1991); Meyaard et al., Science 257:217-219 (1992)), and conditions where excessive cell death results in tissue damage, such as neural injury (Choi, Neuron 1:623-634 (1988)). Neural and immune system development also display important programmed cell death events. In insects, some cell death is observed early, during neuroblast delamination in the formation of the ventral nerve cord (Doe et al., Dev. Biol. 111:193-205 (1985); Jimxc3xa8nez and Campos-Ortega, Neuron 5:81-89 (1990)). In vertebrates, cell death that is not associated with target innervation is observed in the spinal ganglia (Hamburger et al., supra; Pannese, Neuropathol. Appl. Neurobiol. 2:247-267 (1976); Hamburger et al., J. Neurosci. 1:60-71 (1981); Carr et al., Dev. Brain Res. 2:157-162 (1982)). Administration in vivo of nerve growth factor prevents nerve cell death (Hamburger et al., supra), suggesting that competition for factors may participate in the selection of neural cells even at early developmental stages. Progenitor cells of the oligodendrocyte lineage also require certain levels of survival factors during early development (Barres et al., Cell 70:31-46 (1992)). The types of factors that influence cell survival may change as the cells mature (Barres et al., supra), suggesting that different signals may be involved in the selection of cells at different developmental stages. In the immune system, the elimination of cells through cell death functions at multiple stages. T cell maturation may involve two types of selection processes. Cells lacking appropriate receptors fail to be positively selected for further differentiation and are eliminated (Sha et al., Nature 336:73-76 (1988); Teh et al., Nature 335:229-233 (1988)). Cells that do develop receptors but are self-reactive are also eliminated (Kappler et al., Nature 332:35-40 (1988); MacDonald et al., Nature 332:40-45 (1988); Smith et al., Nature 337:181-184 (1989). Regulated cell death thus appears to function together with selection to sculpt an appropriate repertoire of cells.
Programmed cell death typically occurs in conjunction with critical differentiation events. Recent work suggests that programmed cell death is a default fate that will occur unless actively inhibited (Barres et al., supra; Raff, Nature 356:397-400 (1992). In addition, studies done in C. elegans imply that the differentiation pathway and the cell death pathway may be uncoupled genetically (Ellis et al., Cell 44:817-829 (1986); Hengartner et al., Nature 356:495-499 (1992). Genes that function in the cell death pathway have been identified, such as the ced genes of Caenorhabditis elegans (Ellis et al., supra; Hengartner et al., supra). However, genes are also needed to determine when during development that pathway is activated. To coordinate differentiation and death, the activities of genes involved in select differentiation events presumably impinge on control of genes of the death pathway to repress the suicide of appropriate cells.
The Drosophila eye is an excellent genetic system for approaching the problem of how differentiation events and cell death interplay to achieve proper cellular development (Ready, Trends Neurosci. 12:102-110 (1989); Banerjee et al., Neuron 4:177-187 (1990); Rubin, Trends Genet. 7:372-377 (1991)). The adult eye is composed of some 800 repeated neural units called ommatidia, each containing cell types that include three photoreceptor classes, three kinds of pigment cells, cone cells, and a bristle cell complete with socket, neuron, and glial sheath. During the third larval instar, progenitor cells commence differentiation to generate the various cell types (Waddington et al., Proc. Roy. Soc. Lond. (B) 153:155-178 (1960)). Differentiation is marked by a morphogenetic furrow that moves from posterior to anterior across the field of progenitor cells in the eye portion of the eye-antennal imaginal disc (Ready et al., Dev. Biol. 53:217-240 (1976)). Anterior to the morphogenetic furrow, the progenitor cells undergo division to generate an epithelial field for the differentiation events that commence with the furrow. Thus, at a given time, the disc displays a time line of development, the earliest morphologically evident differentiation events being associated with the furrow. Later events occur toward the posterior of the disc, where a pattern emerges of developing cell clusters. Little is known about the events before furrow formation that lead to differentiation, although cell competence, hormones, and possibly inductive interactions appear to be involved (Bodenstein, Postembryotic Development, in Insect Physiology, K. D. Roeder, ed. New York, Wiley and Sons, Inc. pp 822-865 (1953); White, J. Exp. Zool. 148:223-239 (1961); Gateff et al., Roux""s Arch. Dev. Biol. 176:171-189 (1975). Some cell death is a normal part of the developmental process, having been observed in the eye disc during morphogenesis (Fristrom, Mol. Gen. Genet. 103:363-379 (1969); Spreij, Neth. J. Zool. 21:221-264 (1971); Wolff et al., Development 113:825-839 (1991)).
One mutation, eya, has proven useful in the study of eye development. Flies with the eya1 mutation show remarkable specificity for loss of the adult compound eyes (FIGS. 1A and 1B; Sved, Dros Inf. Serv. 63:169 (1986); Renfranz et al., Dev. Biol. 136:411-429 (1989)). All other external structures appear normal, including the adult ocelli, which develop from edges of the eye imaginal discs. In the brain, there is loss of the first optic ganglion (lamina) and reduction in size of the second optic ganglion (medulla), and the lobula and lobula plate show some disorganization (FIGS. 1C and 1D). These brain defects are similar to those observed in other eyeless mutants and are consistent with the influence of retinal neurons on development of the optic lobes (Power, J. Exp. Zool. 94:33-71 (1943); Meyerowitz et al., Dev. Biol. 62:112-142 (1978); Fischbach, Dev. Biol. 91:1-18 (1983); Selleck et al., Neuron 6:83-99 (1991).
In normal development, eye differentiation begins during the third instar larval stage when the morphogenetic furrow sweeps from posterior to anterior across the eye portion of the eye-antennal disc, leaving clusters of differentiating photoreceptor neurons in its wake (Ready et al., supra). The differentiating clusters can be visualized by staining with monoclonal antibodies, such as neuronspecific MAb 22C10 (FIG. 2E; Zipursky et al., Cell 36:15-26 (1984)). In eye1 eye discs, the normal expansion of the eye portion of the disc during the third instar larval stage is arrested, and no furrow forms (FIGS. 2b, 2D, and 2F). When stained with antibodies that normally highlight the differentiating neurons, the mutant eye discs fail to show any evidence of cluster formation (FIG. 2F; also Renfranz et al., supra). In contrast, the antenna portion of the disc expands and differentiates normally. The larval photoreceptor organ also appears normal, and Bolwig""s nerve from the larval visual organ (Bolwig, Vidensk. Medd. fra. Dansk. Naturh. Foren. Bd. 109:80-212 (1946) traverses the eye disc in its path into the optic stalk, as in a normal disc. The eye1 mutation thus appears to affect specifically the progenitor cells that normally form the adult compound eye.
In addition, the study of genes that function in the fly eye has provided insight into the developmental roles of many proteins homologous to human genes (Greenwald et al., Cell 68:271-281 (1992). Anophthalmia, or the lack of eyes, occurs in many organisms, including humans (Apple et al., Ocular Pathology: Clinical applications and Self-Assessment, St. Louis, Mosby-Year Book, Inc. (1991)). In a striking example, congenital aniridia in humans resembles the Small eye mutation of mouse (Glaser et al., Science 250:823-827 (1990), van der Meer-de Jong et al., Genomics 7:270-275 (1990); both result from mutations of genes that are homologous (Hill et al, Nature 354:522-525 (1991); Ton et al., Cell 67:1059-1074 (1991); Jordan et al., Nature Genet. 1:328-332 (1992)).
It is an object of the invention to provide programmed cell death antagonist (PCDA) proteins, and variants thereof, and to produce useful quantities of these PCDA proteins using recombinant DNA techniques.
It is a further object of the invention to provide recombinant nucleic acids encoding PCDA proteins, and expression vectors and host cells containing the nucleic acid encoding the PCDA protein.
An additional object of the invention is to provide methods for producing the PCDA protein, and for regulating the programmed cell death of an organism.
In accordance with the foregoing objects, the present invention provides recombinant PCDA proteins, and isolated or recombinant nucleic acids which encode the PCDA proteins of the present invention. Also provided are expression vectors which comprise DNA encoding a PCDA protein operably linked to transcriptional and translational regulatory DNA, and host cells which contain the expression vectors.
An additional aspect of the present invention provides methods for producing PCDA proteins which comprises culturing a host cell transformed with an expression vector and causing expression of the nucleic acid encoding the PCDA protein to produce a recombinant PCDA protein.
Additionally provided are methods of regulating the programmed cell death of a population of cells in an organism.